[00:03] RAOF: that soon? will ping tseliot about the binary drivers tomorrow then, the november fglrx release isnt out yet and thats the one with 1.11 support [00:04] Sarvatt: Hah. I wasn't aware of that :) [00:04] I would have noticed while preparing stuff (hopfully ☺) [00:08] slangasek, ls -l /dev/input ? [00:09] bryceh: http://paste.ubuntu.com/738776/ [00:09] so was udev updated recently? keyboard/mouse not working bugs have almost always been following a udev update (4-5 times in oneiric..) [00:09] slangasek, interesting! and ls -l /dev/input/by-id/ ? [00:09] http://paste.ubuntu.com/738777/ [00:11] aha [00:12] slangasek, yeah so that's showing events 12 and 13 mapped to your input devices [00:13] hmm [00:13] udev sounds a likely culprit [00:14] aha [00:14] udev ISN'T RUNNING [00:14] score [00:14] slangasek, ahhhh [00:14] :-) [00:15] bryceh: thanks for looking at this :) [00:15] now let me see if I can figure out why it's died, since apport said nothing [00:15] slangasek, sure thing, glad it got sorted [00:15] Setting up udev (175-0ubuntu1) ... [00:15] Installing new version of config file /etc/init/udev.conf ... [00:15] restart: Job failed to restart [00:17] bryceh: X is still not happy loading it, but maybe udev dying has just left it too confused; do you care about debugging that side further, or should I just restart? [00:18] slangasek, go ahead and restart. This is sort of getting into error handling logic which probably is only worth looking into if we get more people encountering this type of problem. [00:19] yep [00:19] thanks again :) [00:19] slangasek, also the X stack is going to get refreshed in the next week or so anyway [00:19] * slangasek nods [00:28] eww, unity updates are "fun", unity-common is built on i386 so it's all uninstallable when amd64 builds and i386 has a huge queue which is a really common scenario [00:29] * slangasek glares at launchpad's redirects, which have handily lost his bug comment due to midair collision with bryceh's bug reassignment [00:30] slangasek, sorry, I bumped it over to udev [00:30] yep [00:31] which is what I was doing :) [00:31] aha :-) [00:50] RAOF: so are you saying you fixed up grabs somehow magically? [00:51] clicking an address bar in a web browser is probably one of the more common things done in ubuntu... [00:53] i'd love to test out anything you've done if you have before 1.11 is pushed to precise is what i'm saying :) [01:00] Sarvatt: No, I haven't fixed grabs yet. [01:01] The "few days max" was the time between everything breaking and everything getting fixed, not the time before I upload 1.11 :) [01:02] planning on uploading 1.11 with broken grabs? [01:02] was talking to chase at UDS, multitouch seriously is almost ready to go in the december timeframe this time around, it might make more sense to wait for that [01:03] 1.11 with backported 1.12 multitouch that is [01:03] s/multitouch/input/ [01:04] No, without broken grabs. [01:05] I was also talking to Chase at UDS, and while I'm optimistic, he certainly wasn't committing to it being done :) [01:05] oh? [01:05] whot is pushing for it soon this time around [01:05] Yeah - they'll be tag-teaming on the implementation. [01:06] cnd was saying he was going to work on the one remaining problem with grabs this time around while whot was on vacation and it should really be ready in december [01:06] But Chase was also “yeah, and we've found out that some things we thought won't actually work properly, and will take a bit more work”… [01:06] Not that it wouldn't be done by December, but more a “sometimes, things happen”. === Guest64961 is now known as LLStarks === soreau_ is now known as soreau [01:20] note to self: dont refuse to look at code because you dont want to get involved with things like unity if things are broken :) [01:21] cant belive i let xorg-edgers be broken for 2 months for 2 lines of code [01:27] 1.5.99 synaptics is unusable either way [01:27] :3 [01:31] on your machine with the upstream code base meaning its just as "broken" on other distros? [01:32] synaptics is heavily patched with the multitouch stuff :( [01:34] what kind of "broken" do you mean? input drivers are the worst, everyone has their own opinions on what they should be [01:35] like acceleration too fast for your tastes? [01:37] many important patches were dropped [01:37] acceleration, sensitivity, and tapping are off-scale [01:38] i could be wrong but those werent changed by any of the patches i dropped [01:38] and that was on top of 1.3.99's hovering and jumpiness [01:38] synaptics had a huge change in sensitivity and crap in the upstream driver [01:39] is there any way to get edgers without an abi-updated synaptics? [01:39] hmm [01:40] because no amount of fiddling has worked [01:40] fiddling with synclient right? [01:40] yeah [01:40] and the guis [01:40] GUI's were only updated in fedora and crap, etfc [01:40] etc [01:41] ok, lemme check [01:42] what is the target synaptic driver for precise? i'd like it to land sooner than later so that this can receive wider testing. [01:42] there is no target, i think i can say that confidently [01:42] we might sync whats in debian [01:42] which is 1.5.0 currently [01:44] as for unity, don't leave things up to ayatana. they don't know drivers or user-oriented ui design. [01:44] theres a few of us that try to keep crap in date with debian wrt syncs and update things in debian but we dont have upload access to debian just git access, 1.5.0 was updated recently so that stcks out as what we might ship with [01:45] the problems you are hitting are well past 1.5.0 though :) [01:45] can i get 1.5 with 1.11 abi? [01:47] you can actually just dget -u -x the debian 1.5.0 .dsc file and debuild -uc -us -b in your current environment [01:48] no multitouch support in edgers so no need to care about the ubuntu delta [01:49] i think a corner tap functionality change might be the only thing you'd hit :) [01:50] the multitouch is unfortunate as only a hover by a second finger will cause a jump [01:50] *the lack [01:54] imho, an epoch update might be in order since functionality is lacking upstream. hopefully 1.5.0 is better. [01:55] brb [01:57] much better, indeed [01:58] 1.5.0? [01:58] yeah, lemme reset the synclient settings though [02:01] i dont really want to put known-working versions in the ppa if possible, rather people complain its broken to upstream if its really broken upstream so they dont release the crap broken :( [02:04] i've been complaining upstream [02:05] people just tune it out and ignore the bug reports with the standard lalalallalal can't hear you [02:05] i think the post 1.5.0 updates might be reliant on gnome-control-center updates going into 3.4 though [02:05] (mouse control panel in gnome-control-center) [02:55] heh, guess i should have quantified "working", cant get the dash up an hour into using unity on 1.11 [02:57] no clue where it's going wrong, only thing in ~/.xsession-errors is a few thousand nautilus errors [03:01] Sarvatt, what do the nautilus errors look like? [03:04] (nautilus:13888): GLib-GObject-CRITICAL **: g_value_get_object: assertion `G_VALUE_HOLDS_OBJECT (value)' failed [03:04] Nautilus-Share-Message: Called "net usershare info" but it failed: 'net usershare' returned error 255: net usershare: cannot open usershare directory /var/lib/samba/usershares. Error No such file or directory [03:04] Please ask your system administrator to enable user sharing. [03:04] minus the last two lines, that only comes up every few hundred iterations [07:46] bryceh: Looks like we've got a couple of blueprints that need bedding down - https://blueprints.launchpad.net/ubuntu/+spec/desktop-p-hybrid-graphics and https://blueprints.launchpad.net/ubuntu/+spec/foundations-p-xorg-lts-updates [07:47] And by ‘bedded down’ mostly I mean ‘needs a victim for the work items’. [07:56] i should probably clean up the hg one [07:57] oh, updated already [08:06] renamed packages? ugh.. [08:23] hmm, so i guess upgrading from .3 to Q would be a no-go [08:23] which would be a downgrade for x [08:24] and the kernel [08:24] so the only upgrade path for >.2 would be to the next lts === soreau_ is now known as soreau === yofel_ is now known as yofel [17:18] awesome http://cgit.freedesktop.org/~eugeni/xf86-video-intel/ [17:18] sna via an xorg.conf option [17:50] RAOF, I've added myself to a couple items; I'll see where I'm at after I draft the MM blueprints; feel free to add me to tasks that you think need my attention [18:09] mdeslaur: any chance you could try http://kernel.ubuntu.com/~kernel-ppa/mainline/daily/current/ to see if the blurry screen problem still exists upstream? jbarnes pinged me saying it should be fixed [18:10] Sarvatt: the ati blurry problem I have or the -intel blurry problem I have? [18:11] intel on your t510 [18:11] https://bugs.freedesktop.org/show_bug.cgi?id=40172 [18:11] Freedesktop bug 40172 in DRM/Intel "[arrandale] Fuzzy screen after dpms cycles on lenovo t510 [bisected, 3.0+]" [Normal,New: ] [18:12] Sarvatt: ah! cool, yeah, I'll give it a try later on. thanks! [18:12] can look into getting it into oneiric if it is [18:12] mdeslaur: appreciate it, we never could find any other machine it reproduced on [18:12] Sarvatt: hopefully it'll be fixed, as it's a PITA :P [20:38] Sarvatt: unfortunately, the problem persists with 3.2.0-999 [20:39] mdeslaur: thanks a ton for checking that, sorry to hear it :( [20:40] Sarvatt: I'm getting used to doing a dpms cycle every few reboots :P [20:57] bryceh: so some of those previous work items you pasted that were postponed i did a long time ago, should I just delete them from the list? [20:57] nvidia/fglrx autoloading stuff [20:59] Sarvatt, just tell me which they were and I'll mark them [20:59] I'm doing a bunch of edits at the moment [21:01] [sarvatt] Fix logic in patches 104 and 105 to correctly select -nvidia when appropriate: DONE [21:01] [sarvatt] Ensure fallbacks work if nvidia/fglrx is failing to load: DONE [21:03] * Sarvatt is knee deep in poulsbo 2.0 aka cedarview "fun", ugh [21:05] got it thanks [22:52] hooray, I can build and install upstream X and it all just works :) [22:58] cnd: what changed? [22:58] ripping geis out of unity? :P [23:00] master is working good here too after doing that [23:01] (xserver) [23:01] Hi [23:02] I need help installing ATI fglrx driver for HP ProBook 4530s laptop with switchable graphics [23:03] Sarvatt, last year about this time I tried to do the same [23:03] I think we were patching in the no-background option [23:04] something like that [23:04] that and maybe gdm wasn't playing nice [23:05] That'd be right, yes. [23:05] I was actually quite amazed I got it all working given how complex it is to configure and install everything [23:06] getting the right --libdir values and all the drivers I needed [23:06] So, how's XI 2.2 going? :) [23:06] RAOF, well, I just got whot's branches of multitouch up and running :) [23:06] for the next two weeks I'll be working on implementing the pointer emulation [23:06] but the basic touch stuff looks to be working well [23:07] That sounds like early December it'd be reasonable to have it in Precise? [23:08] might be [23:08] but that's if everything goes really well [23:08] we have to get it implemented for touchscreens [23:09] and then get it working for trackpads [23:09] so I think it might be more mid-december [23:10] and only if we want to push crack into precise, which the desktop team may not like [23:10] we'll have to figure out the details [23:11] Ok. [23:11] Thanks. [23:13] Bah! Why is the t420s' touchpad so rubbish?